[net.math] estimating p-values

anthony@utcsstat.UUCP (Anthony Ayiomamitis) (08/01/85)

	Is it possible to write a routine for a microcomputer that will
generate a p-value for a given statistic (such as a z, t, F, chi-square)
and degrees of freedom.
	Such a routine would be extremely useful for a program that
performs certain statistical procedures (for example, regression) - in
addition to supplying a test statistic, a program can provide a p-value
as well (even if it is an estimate).

-- 

       	{allegra,ihnp4,linus,decvax}!utzoo!utcsstat!anthony
        {ihnp4|decvax|utzoo|utcsrgv}!utcs!utzoo!utcsstat!anthony

perlman@wanginst.UUCP (Gary Perlman) (08/07/85)

My UNIX|STAT package has a function that gives the probability
of an F ratio.  The code is written in C and should compile
on any PC with a C compiler.  Until now, the package has only
been ported to the IBM PC with the Lattice C compiler.  You can
get the package for a small distribution fee.  Write to me for
more information.
-- 
Gary Perlman  Wang Institute  Tyngsboro, MA 01879  (617) 649-9731
UUCP: decvax!wanginst!perlman             CSNET: perlman@wanginst